DOCSLIB.ORG

The Space D(U) Is Not Br-Complete Annales De L’Institut Fourier, Tome 27, No 4 (1977), P

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
The Space D(U) Is Not Br-Complete Annales De L’Institut Fourier, Tome 27, No 4 (1977), P ANNALES DE L’INSTITUT FOURIER MANUEL VALDIVIA The space D(U) is not Br-complete Annales de l’institut Fourier, tome 27, no 4 (1977), p. 29-43 <http://www.numdam.org/item?id=AIF_1977__27_4_29_0> © Annales de l’institut Fourier, 1977, tous droits réservés. L’accès aux archives de la revue « Annales de l’institut Fourier » (http://annalif.ujf-grenoble.fr/) implique l’accord avec les conditions gé- nérales d’utilisation (http://www.numdam.org/conditions). Toute utilisa- tion commerciale ou impression systématique est constitutive d’une in- fraction pénale. Toute copie ou impression de ce fichier doit conte- nir la présente mention de copyright. Article numérisé dans le cadre du programme Numérisation de documents anciens mathématiques http://www.numdam.org/ Ann. Inst. Fourier, Grenoble 27, 4 (1977), 29-43. THE SPACE ^(Q) IS NOT B.-COMPLETE by Manuel VALDIVIA The purpose of this paper is to study certain classes of locally convex spaces which have non-complete separated quotients. Consequently, we obtain some results about B^- completeness. In particular, we prove that the L. Schwartz space ^(ii) is not B^-complete, where 0, is a non-empty open subset of R7". The vector spaces which are used here are defined on the field of the real or complex numbers K. We shorten « Haus- dorff locally convex space » to « space )>. If <G, H> is a dual pair, we denote by a{G, H) and [i(G, H) the weak topology and the Mackey topology on G, respectively. Sometimes we symbolize by L[^'] a space L with the topology ^. We denote by L' the topological dual of L. If L^ are copies of L, n = 1, 2, . ., we denote by L^ the 00 00 product JJ L^ and by L^ the direct sum © L^. If A n=l n.==-l is an absolutely convex bounded and closed subset of L, LA is the linear hull of A with the locally convex topology which has as fundamental system of neighbourhoods of the . ( 1 ) origin the family ^—A: n = 1, 2, ...^ We say that a n sequence (x^) in L is a Cauchy (convergent) sequence in the sense of Mackey, if there exists an absolutely convex closed and bounded subset B of L, such that (x^) is a Cauchy (convergent) sequence in LB. Then it is natural to define a locally complete space L as a space in which every Cauchy sequence in the sense of Mackey is convergent in the sense of Mackey in L. If a subset M of L is such that for every 30 MANUEL VALDIVIA point x of L there exists a sequence (^) in M which converges to x in the sense of Mackey, M is called locally dense in L. L is called a dual locally complete space if L'[cr(L', L)] is locally complete. If L is topologically iso- morphic to P write L ^ P. Let Q. be a non-empty open subset of the m-dimensional euclidean space R/". Then ^(0.) and ^'(ti) are the L. Schwartz spaces with the strong topologies. By <o we denote the product of countably many one-dimensional spaces. Let s be the vector space of all the rapidly decreasing se- quences, with the metric topology. The dual of s with the Mackey topology is denoted by 5'. A space T is called B^-complete if every dense subspace M of T'[(T(T',T)] coincides with T' when M n A is a(T', T)- closed in A for every equicontinuous set A that belongs to T. Throughout this paper, E denotes an infinite-dimensional Frechet space, different from co, and F a Frechet space which is not Banach. In order to prove Theorem 1, we need the following result, which we gave in [8] : a) Let G be an infinite-dimensional space such that in G'[(?(G', G)] there is an equicontinuous total sequence. Let H be a space with a separable absolutely convex weakly compact total subset. If H is infinite-dimensional, there is a linear mapping u continuous and infective from G into H, such that u(G) is separable, dense in H and u(G)7^H. In Theorem 1, (EJ is a sequence of infinite-dimensional spaces such that in E^ there is a weakly compact absolutely convex and separable subset which is total in E^, n = 1, 2, ... THEOREM 1. — Let L = (E 0 F) X 0 E^. Then there is a n==l quotient of L[(r(L', L)] which is topologically isomorphic to a locally dense and non-closed subspace of JJ E^[cr(E^, E^)]. 71=1 Proof. — Since F is a Frechet space which is not Banach, there exists a closed subspace G of F such that F/G ^ co, (see [21, remark at the end of § 31, 4.(1), p. 432). Let f be the canonical mapping from F onto F/G and let J be the THE SPACE 2(0.) IS NOT By-COMPLETE 31 identity mapping on E. The tensorial product J (g) f is a topological homomorphism from E § F onto E § (F/G) ^ E § (o, (see [I], Chapter I p. 38). Further, if H, is an one-dimensional space, then E § (o = E § (fl H,) ^ n (E § HJ ^ E^, \n=l / n=-l (see [I], Chapter I p. 46). Since E is different from CD, in E'[<i(E', E)] there is an equicontinuous sequence whose closed linear hull M is infinite dimensional. Let M-1- be the orthogonal subspace of E to M, set T = E/M^, and let h be the canonical mappinw from E1* onto E^M^ ^ (E/M1)1' = T\ te Now we use the result a) and be obtain a linear continuous and injective mapping »„, from T into E,, such that u,(T) is dense in E, and M,(T) ^ E,. We define a mapping g from TN X © E, into fj E, such that if "=l n=l x= {xi, a-2, . .., a;,, ...) eT" oo V= (2/1,2/2, •..,yn, ...) e © E,, then n=l g{x, y) = (ui(^) + y^ u^) +^, . ., u,(^) + 2/,, . .). The mapping g is a weak homomorphism from TN X © E^ into IlEn. such that g^ X © E,) is locally dens^and n=:l oo \ n=l / non-closed in JJ E^ (see the proof of Theorem 2 in [7]). n=l oo If I is the identity mapping of 0 E,, then it suffices to take the quotient of L[o(L, L')]by the kernel of the mapping g ° (A ° (J ® /•) x I) in order to obtain the conclu- sion of the theorem, a e d COROLLARY 1.1. — Let (EJ be a sequence of separable infinite-dimensional Frechet spaces. Then there is a quotient of 32 MANUEL VALDIVIA 00 s X © E^, topologically isomorphic to a non-closed dense 71=1 oo subspace of JJ E^. 71=1 Proof. — Since E^ is a separable Frechet space there exists an absolutely convex compact separable subset of E^, which is total in E^. We put T= {^y)eR^:x2+y2=l}. Then ^(T) ^ s and ^(T X T) ^ 5, (see [I], Chapter II, p. 54). Since ^(T) § ^(T) ^ ^(T X T), (see [I], Chapter II, p. 84-85), we have that s S s ^ s. On the other hand, s is an infinite-dimensional Frechet space, different from co, which is not Banach. 00 Using Theorem 1 we obtain a subspace P of s X ® E^ / oo \ / ^ ^ n=i such that [s X (B E^)/P is weakly isomorphic to a dense \ 7l=l // 00 00 non-closed subspace of JJ E^. Since (/^ X ® E^)/\ / P is a T»=I \ n=l // 00 Mackey space and every subspace of ]J[ E^ is also a Mackey 71=1 space, we have that (/ s X ©°° E/,\) // P is topologically iso- \ n=l II oo morphic to a dense non-closed subspace of TT E^ q.e.d. 71=1 Note 1. — 0. G. Smoljanov proves in [6] that there is a closed subspace H in s X o/^ such that {s X o^/H is a non closed dense subspace of co. This result is a conse- quence of our Corollary 1.1, identifing o with E^, n = 1, 2, ..., and noticing that c^ ^ G). THEOREM 2. — Let (EJ fee a sequence of non-Banach 00 Frechet spaces. Then (E § F) X © E^ /ia5 a quotient which is 71=1 topologically isomorphic to a non-closed dense subspace of co. Proof. — Since E^ is not a Banach space there is a subspace G, of (E § F) X © E, lying in E,, such that EJG, ^ co. 71=1 THE SPACE ^(Q) IS NOT Bp-COMPLETE 33 Let G = © G,. Then 71=1 f(E ® F) x © E,1/G ^ (E ® F) x © (EJG,) ^ (E ® F) x (^ L n==l -i/ 71=1 and the conclusion follows straightforward from Theorem 1 applied to (E § F) X co^. q.e.d. COROLLARY 1.2. — There is a quotient of s^ which is topologically isomorphic to a non-closed dense subspace of (o. Proof. — Since s § s ^ s, it is possible to write (s 0 s) X s^ ^ s^\ Now we apply Theorem 2. q.e.d. LEMMA. — The space ^(Q) has a closed subspace topolo- gically isomorphic to s^\ Proof. — In 0, let (K^) be a sequence of compact sets such that K^ c: K^+i, n == 1, 2, . ., being K^+i the interior 00 set of K^+i, and [_j K^ == ^2. Suppose that Ki ^ 0. 71=1 We put Ko = 0. Let R^ be a closed m-dimensional rectangle containded in K^ - K^_i, n = 1, 2, ..
Load more

Recommended publications
  • An Abstract Algebraic–Topological Approach to the Notions of a First
    An abstract algebraic{topological approach to the notions of a ¯rst and a second dual space I H. Poppe, R. Bartsch 1 Introduction By IR and C= we denote the reals and the complex numbers respectively and by IK we mean IR or C= . Now let (X; jj ¢ jj) be a normed IK{vector space and X0 := ff : X ! IKj f linear and continuousg the (¯rst) dual space (dual) of X. Hence X0 consists of functions which at the same time are an algebraic homomorphism (a linear map) and a topological homomorphism (a continuous map). By the usual operatornorm, (X0; jj ¢ jj) again becomes a normed space (even a Banach{space) meaning that X and X0 belong to the same class of spaces. Hence at once we can construct the second dual space (bidual) X00 by: X00 := ((X; jj ¢ jj)0; jj ¢ jj)0. Then the canonical map J : X ! X00 is de¯ned via the evaluation map(1) ! : X £ IKX ! IK by J(x) := !(x; ¢) 2 X00 with 8x 2 X : !(x; ¢): X0 ! IK; 8x0 2 X0 : !(x; ¢)(x0) = !(x; x0) = x0(x). 1.1 Proposition X !(x; ¢):(IK ;¿p) ! IK is continuous for all x 2 X. X X ¿p Proof: For any net (fi)i2I from IK , f 2 IK , fi ¡! f implies especially fi(x) ! f(x) and thus !(x; fi) = fi(x) ! f(x) = !(x; f). 1.2 Remark So the map J is well{de¯ned since we have 8x 2 X : !(x; ¢) 2 X00, i.e. !(x; ¢): 0 (X ; jj ¢ jj) ! IK is linear and continuous (w.r.t.
    [Show full text]
  • A Study of Uniform Boundedness
    CANADIAN THESES H pap^^ are missmg, contact the university whch granted the Sil manque des pages, veuiflez communlquer avec I'univer- degree.-. sit6 qui a mf&B b grade. henaPsty coWngMed materiats (journal arbcks. puMished Les documents qui tont c&j& robjet dun dfoit dautew (Wicks tests, etc.) are not filmed. de revue, examens pubiles, etc.) ne scmt pas mlcrofiimes. Repockbm in full or in part of this film is governed by the lareproduction, m&ne partielk, de ce microfibn est soumise . Gawc&n CoWnght Act, R.S.C. 1970, c. G30. P!ease read B h Ldcanaden ne sur b droit &auteur, SRC 1970, c. C30. the alrhmfm fmwhich mpanythfs tksts. Veuikz pm&e des fmhdautortsetion qui A ac~ompegnentcette Mse. - - - -- - - -- - -- ppp- -- - THIS DISSERTATlON LA W&E A - - -- EXACTLY AS RECEIVED NOUS L'AVONS RECUE d the fitsr. da prdrer w de vervke des exsaplairas du fib, *i= -4 wit- tM arrhor's rrrittm lsrmissitn. (Y .~-rmntrep&ds sns rrror;s~m&rite de /.auefa. e-zEEEsXs-a--==-- , THE RZ~lREHEHTFOR THE DEGREB OF in the Wpartesent R.T. SBantunga, 1983 SLW PRllSER BNFJERSI'PY Decenber 1983 Hae: Degree: Title of Thesis: _ C, Godsil ~srtsrnalgnamniner I3e-t of Hathematics Shmn Praser University 'i hereby grant to Simn Fraser University tba right to .lend ,y 'hcsis, or ~.ndadessay 4th title of which is 5taun helor) f to users ot the 5imFrasar University Library, and to mke partial or r'H.lgtrt copies oniy for such users 5r in response to a request fram the 1 Ibrary oh any other university, or other educationat institution, on its orn behaif or for one of its users.
    [Show full text]
  • International Centre for Theoretical Physics
    zrrl^^^'C^z ic/89/30i INTERNATIONAL CENTRE FOR THEORETICAL PHYSICS ON THE APPROXIMATIVE NORMAL VALUES OF MULTIVALUES OPERATORS IN TOPOLOGICAL VECTOR SPACE Nguyen Minh Chuong and INTERNATIONAL ATOMIC ENERGY Khuat van Ninh AGENCY UNITED NATIONS EDUCATIONAL, SCIENTIFIC AND CULTURAL ORGANIZATION IC/89/301 SUNTO International Atomic Energy Agency and In questo articolo si considera il problemadell'approssimaztone di valori normali di oper- United Nations Educational Scientific and Cultural Organization atori chiusi lineari multivoci dallo spazio vettoriale topologico de Mackey, a valori in un E-spazio. INTERNATIONAL CENTRE FOR THEORETICAL PHYSICS Si dimostrano l'esistenza di un valore normale e la convergenza dei valori approssimanti il valore normale. ON THE APPROXIMATIVE NORMAL VALUES OF MULTIVALUED OPERATORS IN TOPOLOGICAL VECTOR SPACE * Nguyen Minn Chuong * International Centre for Theoretical Physics, Trieste, Italy. and Khuat van Ninh Department of Numerical Analysis, Institute of Mathematics, P.O. Box 631, Bo Ho, 10 000 Hanoi, Vietnam. ABSTRACT In this paper the problem of approximation of normal values of multivalued linear closed operators from topological vector Mackey space into E-space is considered. Existence of normal value and covergence of approximative values to normal value are proved. MIRAMARE - TRIESTE September 1989 To be submitted for publication. Permanent address: Department of Numerical Analysis, Institute of Mathematics, P.O. Box 631, Bo Ho 10 000 Hanoi, Vietnam. 1. Introduction Let X , Y be closed Eubspaces of X,Y respectively, where X is normed In recent years problems on normal value and on approximation of space and Y is Hackey space. Let T be a linear multivalued operator from normal value of multivalued linear operators have been investigated Y to X .
    [Show full text]
  • Let H Be a Hilbert Space. on B(H), There Is a Whole Zoo of Topologies
    Let H be a Hilbert space. On B(H), there is a whole zoo of topologies weaker than the norm topology – and all of them are considered when it comes to von Neumann algebras. It is, however, a good idea to concentrate on one of them right from the definition. My choice – and Murphy’s [Mur90, Chapter 4] – is the strong (or strong operator=STOP) topology: Definition. A von Neumann algebra is a ∗–subalgebra A ⊂ B(H) of operators acting nonde- generately(!) on a Hilbert space H that is strongly closed in B(H). (Every norm convergent sequence converges strongly, so A is a C∗–algebra.) This does not mean that one has not to know the other topologies; on the contrary, one has to know them very well, too. But it does mean that proof techniques are focused on the strong topology; if we use a different topology to prove something, then we do this only if there is a specific reason for doing so. One reason why it is not sufficient to worry only about the strong topology, is that the strong topology (unlike the norm topology of a C∗–algebra) is not determined by the algebraic structure alone: There are “good” algebraic isomorphisms between von Neumann algebras that do not respect their strong topologies. A striking feature of the strong topology on B(H) is that B(H) is order complete: Theorem (Vigier). If aλ λ2Λ is an increasing self-adjoint net in B(H) and bounded above (9c 2 B(H): aλ ≤ c8λ), then aλ converges strongly in B(H), obviously to its least upper bound in B(H).
    [Show full text]
  • Representations of a Compact Group on a Banach Space
    Journal of the Mathematical Society of Japan Vol. 7, No. 3, June, 1955 Representations of a compact group on a Banach space. By Koji SHIGA (Received Jan. 14, 1955) Introduction. Let G be a compact group and IYIa complex normed linear space. We mean by a representation {, T(a)} a homomorphism of G into the group of bounded linear operators on such that a --~ T(a). The following types of representations are important : i) algebraic repre- sentation, ii) bounded algebraic representation,l' iii) weakly measurable representation,2' iv) strongly measurable representation,3' v) weakly continuous representation,4' vi) strongly continuous representation.5' Clearly each condition above becomes gradually stronger when it runs from i) to vi), except iv). When the representation space 1 of {IJI,T(a)} is a Banach space, it is called a B-representation. In considering a B-representation {8, T(a)}, what we shall call the conjugate representation {8*, T *(a)} arouses our special interest. For example, we can prove, by considering {**, T **(a)} of a weakly continuous B-representation {8, T(a)}, the equivalence of all four condi- tions (Theorem 2) : 1) {l3, T(a)} is completely decomposable, 2) {8, T(a)} is strongly continuous, 3) {3, T(a)} is strongly measurable, 4) for any given x e 3, the closed invariant subspace generated by T(a)x, 1) Algebraicreps esentation such that IIT(a) II< M for a certain M(> 0). 2) Boundedalgebraic representation such that f(T(a)x) is a measurable function on G for any xe3 and f€3,*where 9* meansthe conjugatespace of .
    [Show full text]
  • On Hereditary Reflexivity of Topological Vector Spaces
    GLASNIK MATEMATICKIˇ Vol. 43(63)(2008), 397 – 422 ON HEREDITARY REFLEXIVITY OF TOPOLOGICAL VECTOR SPACES. THE DE RHAM COCHAIN AND CHAIN SPACES Ju. T. Lisica Peoples’ Friendship University of Russia, Russia Dedicated to Professor Sibe Mardeˇsi´c Abstract. For proving reflexivity of the spaces of de Rham coho- mology and homology of C∞-manifolds the author considers the notion of hereditary reflexivity as well as the notion of dual hereditary reflexivity of locally convex topological vector spaces which is interesting in itself. Com- plete barrelled nuclear spaces with complete nuclear duals turn out to be hereditarily reflexive. The Pontryagin duality in locally convex topological spaces is also considered. 1. Introduction p p c Let Ω (M), Ωc (M), Ωp(M), Ωp(M), p ≥ 0, be the de Rham cochain com- plex of C∞-differential forms, the de Rham cochain complex of C∞-differential forms with compact supports and the de Rham chain complex of C∞-differen- tial forms with coefficients in Schwartz distributions with compact sup- ports (currents with compact supports), the de Rham chain complex of C∞-differential forms with coefficients in Schwartz distributions (currents) of a C∞-manifold M, respectively. The locally convex topology on Ωp(M) is the topology of compact con- p vergence for all derivatives, on Ωc (M) it is the topology of the strict inductive 2000 Mathematics Subject Classification. 46A03, 46A04, 55N07, 55P55. Key words and phrases. De Rham cohomology, currents, nuclear spaces, hereditary reflexivity, dual hereditary reflexivity, Pontryagin duality. This research was supported by a grant of Russia’s Fund of Fundamental Researches (Grant # 06-01-00341).
    [Show full text]
  • A General Form of Gelfand-Kazhdan Criterion
    A GENERAL FORM OF GELFAND-KAZHDAN CRITERION BINYONG SUN AND CHEN-BO ZHU Abstract. We formalize the notion of matrix coefficients for distributional vectors in a representation of a real reductive group, which consist of generalized functions on the group. As an application, we state and prove a Gelfand-Kazhdan criterion for a real reductive group in very general settings. 1. Tempered generalized functions and Casselman-Wallach representations In this section, we review some basic terminologies in representation theory, which are necessary for this article. The two main ones are tempered generalized functions and Casselman-Wallach representations. We refer the readers to [W1, W2] as general references. Let G be a real reductive Lie group, by which we mean that (a) the Lie algebra g of G is reductive; (b) G has finitely many connected components; and (c) the connected Lie subgroup of G with Lie algebra [g, g] has a finite center. We say that a (complex valued) function f on G is of moderate growth if there is a continuous group homomorphism ρ : G → GLn(C), for some n ≥ 1, such that t t |f(x)| ≤ tr(ρ(x) ρ(x)) + tr(ρ(x−1) ρ(x−1)), x ∈ G. Here “¯” stands for the complex conjugation, and “ t” the transpose, of a matrix. ∞ arXiv:0903.1409v6 [math.RT] 25 Jan 2011 A smooth function f ∈ C (G) is said to be tempered if Xf has moderate growth for all X in the universal enveloping algebra U(gC). Here and as usual, gC is the complexification of g, and U(gC) is identified with the space of all left invariant differential operators on G.
    [Show full text]
  • Nonlinear Spectral Theory Henning Wunderlich
    Nonlinear Spectral Theory Henning Wunderlich Dr. Henning Wunderlich, Frankfurt, Germany E-mail address: [email protected] 1991 Mathematics Subject Classification. 46-XX,47-XX I would like to sincerely thank Prof. Dr. Delio Mugnolo for his valuable advice and support during the preparation of this thesis. Contents Introduction 5 Chapter 1. Spaces 7 1. Topological Spaces 7 2. Uniform Spaces 11 3. Metric Spaces 14 4. Vector Spaces 15 5. Topological Vector Spaces 18 6. Locally-Convex Spaces 21 7. Bornological Spaces 23 8. Barreled Spaces 23 9. Metric Vector Spaces 29 10. Normed Vector Spaces 30 11. Inner-Product Vector Spaces 31 12. Examples 31 Chapter 2. Fixed Points 39 1. Schauder-Tychonoff 39 2. Monotonic Operators 43 3. Dugundji and Quasi-Extensions 51 4. Measures of Noncompactness 53 5. Michael Selection 58 Chapter 3. Existing Spectra 59 1. Linear Spectrum and Properties 59 2. Spectra Under Consideration 63 3. Restriction to Linear Operators 76 4. Nonemptyness 77 5. Closedness 78 6. Boundedness 82 7. Upper Semicontinuity 84 Chapter 4. Applications 87 1. Nemyckii Operator 87 2. p-Laplace Operator 88 3. Navier-Stokes Equations 92 Bibliography 97 Index 101 3 Introduction The term Spectral Theory was coined by David Hilbert in his studies of qua- dratic forms in infinitely-many variables. This theory evolved into a beautiful blend of Linear Algebra and Analysis, with striking applications in different fields of sci- ence. For example, the formulation of the calculus of Quantum Mechanics (e.g., POVM) would not have been possible without such a (Linear) Spectral Theory at hand.
    [Show full text]
  • Topological Vector Spaces
    Chapter 2 Topological Vector Spaces 2.1 Definition and properties of a topological vector space In this section we are going to consider vector spaces over the field K of real or complex numbers which is given the usual euclidean topology defined by means of the modulus. Definition 2.1.1. A vector space X over K is called a topological vector space (t.v.s.) if X is provided with a topology ⌧ which is compatible with the vector space structure of X, i.e. ⌧ makes the vector space operations both continuous. More precisely, the condition in the definition of t.v.s. requires that: X X X ⇥ ! (x, y) x + y vector addition 7! K X X ⇥ ! (λ, x) λx scalar multiplication 7! are both continuous when we endow X with the topology ⌧, K with the eu- clidean topology, X X and K X with the correspondent product topologies. ⇥ ⇥ Remark 2.1.2. If (X,⌧) is a t.v.s then it is clear from Definition 2.1.1 that N (n) (n) N k=1 λk xk k=1 λkxk as n w.r.t. ⌧ if for each k =1,...,N ! (n) !1 as n we have that λ λk w.r.t. the euclidean topology on K and P !1 P k ! x(n) x w.r.t. ⌧. k ! k Let us discuss now some examples and counterexamples of t.v.s. Examples 2.1.3. a) Every vector space X over K endowed with the trivial topology is a t.v.s.. 15 2. Topological Vector Spaces b) Every normed vector space endowed with the topology given by the metric induced by the norm is a t.v.s.
    [Show full text]
  • Subspaces and Quotients of Topological and Ordered Vector Spaces
    Zoran Kadelburg Stojan Radenovi´c SUBSPACES AND QUOTIENTS OF TOPOLOGICAL AND ORDERED VECTOR SPACES Novi Sad, 1997. CONTENTS INTRODUCTION::::::::::::::::::::::::::::::::::::::::::::::::::::: 1 I: TOPOLOGICAL VECTOR SPACES::::::::::::::::::::::::::::::: 3 1.1. Some properties of subsets of vector spaces ::::::::::::::::::::::: 3 1.2. Topological vector spaces::::::::::::::::::::::::::::::::::::::::: 6 1.3. Locally convex spaces :::::::::::::::::::::::::::::::::::::::::::: 12 1.4. Inductive and projective topologies ::::::::::::::::::::::::::::::: 15 1.5. Topologies of uniform convergence. The Banach-Steinhaus theorem 21 1.6. Duality theory ::::::::::::::::::::::::::::::::::::::::::::::::::: 28 II: SUBSPACES AND QUOTIENTS OF TOPOLOGICAL VECTOR SPACES ::::::::::::::::::::::::::::::::::::::::::::::::::::::::: 39 2.1. Subspaces of lcs’s belonging to the basic classes ::::::::::::::::::: 39 2.2. Subspaces of lcs’s from some other classes :::::::::::::::::::::::: 47 2.3. Subspaces of topological vector spaces :::::::::::::::::::::::::::: 56 2.4. Three-space-problem for topological vector spaces::::::::::::::::: 60 2.5. Three-space-problem in Fr´echet spaces:::::::::::::::::::::::::::: 65 III: ORDERED TOPOLOGICAL VECTOR SPACES :::::::::::::::: 72 3.1. Basics of the theory of Riesz spaces::::::::::::::::::::::::::::::: 72 3.2. Topological vector Riesz spaces ::::::::::::::::::::::::::::::::::: 79 3.3. The basic classes of locally convex Riesz spaces ::::::::::::::::::: 82 3.4. l-ideals of topological vector Riesz spaces :::::::::::::::::::::::::
    [Show full text]
  • Mackey +0-Barrelled Spaces Stephen A
    Advances in Mathematics 145, 230238 (1999) Article ID aima.1998.1815, available online at http:ÂÂwww.idealibrary.com on Mackey +0-Barrelled Spaces Stephen A. Saxon* Department of Mathematics, University of Florida, P.O. Box 118105, Gainesville, Florida 32611-8105 E-mail: saxonÄmath.ufl.edu and Ian Tweddle* Department of Mathematics, University of Strathclyde, Glasgow G11XH, Scotland, United Kingdom E-mail: i.tweddleÄstrath.ac.uk CORE Metadata, citation and similar papers at core.ac.uk Provided by Elsevier - PublisherReceived Connector March 25, 1998; accepted December 14, 1998 In the context of ``Reinventing weak barrelledness,'' the best possible versions of the RobertsonSaxonRobertson Splitting Theorem and the SaxonTweddle Fit and Flat Components Theorem are obtained by weakening the hypothesis from ``barrelled'' to ``Mackey and +0-barreled.'' An example showing that the latter does not imply the former validates novelty, answers an old question, and completes a robust linear picture of ``Mackey weak barrelledness'' begun several decades ago. 1999 Academic Press Key Words: weak barrelledness; Mackey topology; splitting theorem. 1. INTRODUCTION Barrelled spaces have occupied an important place in the theory of locally convex spaces since its earliest days. This is probably due to the fact that they provide a vehicle for generalizing certain important properties of Banach spaces; for example, they form the class of domain spaces for a natural generalization of Banach's closed graph theorem and, in their dual characterization, they embody the conclusion of the BanachSteinhaus theorem for continuous linear functionals. Unlike Banach spaces, they are closed under the taking of inductive limits, countable-codimensional subspaces, products, etc. A barrel in a locally convex space is a closed * We are grateful for support and hospitality from EPSRC GRÂL67257 and Strathclyde University, and for helpful discussions with Professors J.
    [Show full text]
  • Quasi-Barrelled Locally Convex Spaces 811
    i960] quasi-barrelled locally convex spaces 811 Bibliography 1. R. E. Lane, Absolute convergence of continued fractions, Proc. Amer. Math. Soc. vol. 3 (1952) pp. 904-913. 2. R. E. Lane and H. S. Wall, Continued fractions with absolutely convergent even and odd parts, Trans. Amer. Math. Soc. vol. 67 (1949) pp. 368-380. 3. W. T. Scott and H. S. Wall, A convergence theorem for continued fractions, Trans. Amer. Math. Soc. vol. 47 (1940) pp. 155-172. 4. H. S. Wall, Analytic theory of continued fractions, New York, D. Van Nostrand Company, Inc., 1948. The University of Texas and University of Houston QUASI-BARRELLED LOCALLY CONVEX SPACES MARK MAHOWALD AND GERALD GOULD 1. Introduction and preliminary definitions. The main object of this paper is to answer some problems posed by Dieudonné in his paper Denumerability conditions in locally convex vector spaces [l]. His two main results are as follows: Proposition 1. If Eis a barrelled space on which there is a countable fundamental system of convex compact subsets, [Definition 1.2] then it is the strong dual of a Fréchet-Montel Space. Proposition 2. If E is either bornological or barrelled, and if there is a countable fundamental system of compact subsets, then E is dense in the strong dual of a Fréchet-Montel Space. Two questions raised by Dieudonné in connection with these results are: (a) If E is either bornological or barrelled then it is certainly quasi- barrelled [l, Chapter 3, §2, Example 12]. Can one substitute this weaker condition on E in Proposition 2? (b) Is there is an example of a quasi-barrelled space which is neither barrelled nor bornological? We shall show that the answer to (a) is "Yes," and that the answer to (b) is also "Yes," so that the generalization is in fact a real one.
    [Show full text]